|Looking Inside the Black Box of Machine Learning Methods: Applications in Analytical Chemistry
|Big Data Chemometrics and Method development (In-Silico) (KVCV)
|JMP, SAS Institute
Abstract Information :
Developing robust and accurate analytical methods relies on collecting the best data and extracting maximum insight. The volume, diversity and complexity of data in analytical chemistry is increasing all the time. This means that analytical chemists often need the skills and tools of a data scientist to efficiently and effectively deal with these challenges. There is much promise around Machine Learning methods, in particular. However, many of the methods and their results can be appear to be somewhat of a black box.
In this presentation we will show the insight and efficiency that can be gained from applying modern data analytics to analytical chemistry data. You will also gain a greater understanding of the mechanisms behind methods including Neural Nets, Clustering and Decision Trees. And how you can make sense of the options to find the most useful solution for your analytical problem by visually interacting with the data and the models.